schuetoliver
/
handson-ml
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Sync notebook with book's code examples, and better identify extra code

main
Aurélien Geron 2022-02-19 18:17:36 +13:00
parent 1c2421fc88
commit b63019fd28
9 changed files with 318 additions and 301 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

58
02_end_to_end_machine_learning_project.ipynb
View File

@ -177,7 +177,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – code to save the figures as high-res PNGs for the book\n",
"# extra code – code to save the figures as high-res PNGs for the book\n",
"\n",
"IMAGES_PATH = Path() / \"images\" / \"end_to_end_project\"\n",
"IMAGES_PATH.mkdir(parents=True, exist_ok=True)\n",
@ -197,7 +197,7 @@
"source": [
"import matplotlib.pyplot as plt\n",
"\n",
"# not in the book – the next 5 lines define the default font sizes\n",
"# extra code – the next 5 lines define the default font sizes\n",
"plt.rc('font', size=14)\n",
"plt.rc('axes', labelsize=14, titlesize=14)\n",
"plt.rc('legend', fontsize=14)\n",
@ -205,7 +205,7 @@
"plt.rc('ytick', labelsize=10)\n",
"\n",
"housing.hist(bins=50, figsize=(12, 8))\n",
"save_fig(\"attribute_histogram_plots\") # not in the book\n",
"save_fig(\"attribute_histogram_plots\") # extra code\n",
"plt.show()"
]
},
@ -351,7 +351,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – shows how to compute the 10.7% proba of getting a bad sample\n",
"# extra code – shows how to compute the 10.7% proba of getting a bad sample\n",
"\n",
"from scipy.stats import binom\n",
"\n",
@ -375,7 +375,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – shows another way to estimate the probability of bad sample\n",
"# extra code – shows another way to estimate the probability of bad sample\n",
"\n",
"np.random.seed(42)\n",
"\n",
@ -403,7 +403,7 @@
"housing[\"income_cat\"].value_counts().sort_index().plot.bar(rot=0, grid=True)\n",
"plt.xlabel(\"Income category\")\n",
"plt.ylabel(\"Number of districts\")\n",
"save_fig(\"housing_income_cat_bar_plot\") # not in the book\n",
"save_fig(\"housing_income_cat_bar_plot\") # extra code\n",
"plt.show()"
]
},
@ -464,7 +464,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code computes the data for Figure 2–10\n",
"# extra code – computes the data for Figure 2–10\n",
"\n",
"def income_cat_proportions(data):\n",
" return data[\"income_cat\"].value_counts() / len(data)\n",
@ -524,7 +524,7 @@
"outputs": [],
"source": [
"housing.plot(kind=\"scatter\", x=\"longitude\", y=\"latitude\", grid=True)\n",
"save_fig(\"bad_visualization_plot\") # not in the book\n",
"save_fig(\"bad_visualization_plot\") # extra code\n",
"plt.show()"
]
},
@ -535,7 +535,7 @@
"outputs": [],
"source": [
"housing.plot(kind=\"scatter\", x=\"longitude\", y=\"latitude\", grid=True, alpha=0.2)\n",
"save_fig(\"better_visualization_plot\") # not in the book\n",
"save_fig(\"better_visualization_plot\") # extra code\n",
"plt.show()"
]
},
@ -549,7 +549,7 @@
" s=housing[\"population\"] / 100, label=\"population\",\n",
" c=\"median_house_value\", cmap=\"jet\", colorbar=True,\n",
" legend=True, sharex=False, figsize=(10, 7))\n",
"save_fig(\"housing_prices_scatterplot\") # not in the book\n",
"save_fig(\"housing_prices_scatterplot\") # extra code\n",
"plt.show()"
]
},
@ -573,7 +573,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates the first figure in the chapter\n",
"# extra code – this cell generates the first figure in the chapter\n",
"\n",
"# Download the California image\n",
"filename = \"california.png\"\n",
@ -638,7 +638,7 @@
"attributes = [\"median_house_value\", \"median_income\", \"total_rooms\",\n",
" \"housing_median_age\"]\n",
"scatter_matrix(housing[attributes], figsize=(12, 8))\n",
"save_fig(\"scatter_matrix_plot\") # not in the book\n",
"save_fig(\"scatter_matrix_plot\") # extra code\n",
"plt.show()"
]
},
@ -650,7 +650,7 @@
"source": [
"housing.plot(kind=\"scatter\", x=\"median_income\", y=\"median_house_value\",\n",
" alpha=0.1, grid=True)\n",
"save_fig(\"income_vs_house_value_scatterplot\") # not in the book\n",
"save_fig(\"income_vs_house_value_scatterplot\") # extra code\n",
"plt.show()"
]
},
@ -1195,7 +1195,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates Figure 2–17\n",
"# extra code – this cell generates Figure 2–17\n",
"fig, axs = plt.subplots(1, 2, figsize=(8, 3), sharey=True)\n",
"housing[\"population\"].hist(ax=axs[0], bins=50)\n",
"housing[\"population\"].apply(np.log).hist(ax=axs[1], bins=50)\n",
@ -1219,7 +1219,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code just shows that we get a uniform distribution\n",
"# extra code – just shows that we get a uniform distribution\n",
"percentiles = [np.percentile(housing[\"median_income\"], p)\n",
" for p in range(1, 100)]\n",
"flattened_median_income = pd.cut(housing[\"median_income\"],\n",
@ -1251,7 +1251,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates Figure 2–18\n",
"# extra code – this cell generates Figure 2–18\n",
"\n",
"ages = np.linspace(housing[\"housing_median_age\"].min(),\n",
" housing[\"housing_median_age\"].max(),\n",
@ -1488,7 +1488,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates Figure 2–19\n",
"# extra code – this cell generates Figure 2–19\n",
"\n",
"housing_renamed = housing.rename(columns={\n",
" \"latitude\": \"Latitude\", \"longitude\": \"Longitude\",\n",
@ -1638,7 +1638,7 @@
"metadata": {},
"outputs": [],
"source": [
"df_housing_num_prepared.head(2) # not in the book"
"df_housing_num_prepared.head(2) # extra code"
]
},
{
@ -1737,7 +1737,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code shows that we can get a DataFrame out if we want\n",
"# extra code – shows that we can get a DataFrame out if we want\n",
"housing_prepared_fr = pd.DataFrame(\n",
" housing_prepared,\n",
" columns=preprocessing.get_feature_names_out(),\n",
@ -1866,7 +1866,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code computes the error ratios discussed in the book\n",
"# extra code – computes the error ratios discussed in the book\n",
"error_ratios = housing_predictions[:5].round(-2) / housing_labels.iloc[:5].values - 1\n",
"print(\", \".join([f\"{100 * ratio:.1f}%\" for ratio in error_ratios]))"
]
@ -1942,7 +1942,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code computes the error stats for the linear model\n",
"# extra code – computes the error stats for the linear model\n",
"lin_rmses = -cross_val_score(lin_reg, housing, housing_labels,\n",
" scoring=\"neg_root_mean_squared_error\", cv=10)\n",
"pd.Series(lin_rmses).describe()"
@ -2062,7 +2062,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code shows part of the output of get_params().keys()\n",
"# extra code – shows part of the output of get_params().keys()\n",
"print(str(full_pipeline.get_params().keys())[:1000] + \"...\")"
]
},
@ -2107,7 +2107,7 @@
"cv_res = pd.DataFrame(grid_search.cv_results_)\n",
"cv_res.sort_values(by=\"mean_test_score\", ascending=False, inplace=True)\n",
"\n",
"# not in the book – these few lines of code just make the DataFrame look nicer\n",
"# extra code – these few lines of code just make the DataFrame look nicer\n",
"cv_res = cv_res[[\"param_preprocessing__geo__n_clusters\",\n",
" \"param_random_forest__max_features\", \"split0_test_score\",\n",
" \"split1_test_score\", \"split2_test_score\", \"mean_test_score\"]]\n",
@ -2174,7 +2174,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code displays the random search results\n",
"# extra code – displays the random search results\n",
"cv_res = pd.DataFrame(rnd_search.cv_results_)\n",
"cv_res.sort_values(by=\"mean_test_score\", ascending=False, inplace=True)\n",
"cv_res = cv_res[[\"param_preprocessing__geo__n_clusters\",\n",
@ -2213,7 +2213,7 @@
},
"outputs": [],
"source": [
"# not in the book – plots a few distributions you can use in randomized search\n",
"# extra code – plots a few distributions you can use in randomized search\n",
"\n",
"from scipy.stats import randint, uniform, geom, expon\n",
"\n",
@ -2275,7 +2275,7 @@
},
"outputs": [],
"source": [
"# not in the book – shows the difference between expon and reciprocal\n",
"# extra code – shows the difference between expon and reciprocal\n",
"\n",
"from scipy.stats import reciprocal\n",
"\n",
@ -2410,7 +2410,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – shows how to compute a confidence interval for the RMSE\n",
"# extra code – shows how to compute a confidence interval for the RMSE\n",
"m = len(squared_errors)\n",
"mean = squared_errors.mean()\n",
"tscore = stats.t.ppf((1 + confidence) / 2, df=m - 1)\n",
@ -2431,7 +2431,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – computes a confidence interval again using z-score\n",
"# extra code – computes a confidence interval again using z-score\n",
"zscore = stats.norm.ppf((1 + confidence) / 2)\n",
"zmargin = zscore * squared_errors.std(ddof=1) / np.sqrt(m)\n",
"np.sqrt(mean - zmargin), np.sqrt(mean + zmargin)"
@ -2477,7 +2477,7 @@
"source": [
"import joblib\n",
"\n",
"# not in the book – excluded for conciseness\n",
"# extra code – excluded for conciseness\n",
"from sklearn.cluster import KMeans\n",
"from sklearn.base import BaseEstimator, TransformerMixin\n",
"from sklearn.metrics.pairwise import rbf_kernel\n",

60
03_classification.ipynb
View File

@ -142,7 +142,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – it's a bit too long\n",
"# extra code – it's a bit too long\n",
"print(mnist.DESCR)"
]
},
@ -152,7 +152,7 @@
"metadata": {},
"outputs": [],
"source": [
"mnist.keys() # not in the book – we only use data and target in this notebook"
"mnist.keys() # extra code – we only use data and target in this notebook"
]
},
{
@ -216,7 +216,7 @@
"\n",
"some_digit = X[0]\n",
"plot_digit(some_digit)\n",
"save_fig(\"some_digit_plot\") # not in the book\n",
"save_fig(\"some_digit_plot\") # extra code\n",
"plt.show()"
]
},
@ -235,7 +235,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates Figure 3–2\n",
"# extra code – this cell generates and saves Figure 3–2\n",
"plt.figure(figsize=(9, 9))\n",
"for idx, image_data in enumerate(X[:100]):\n",
" plt.subplot(10, 10, idx + 1)\n",
@ -427,7 +427,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code also computes the precision: TP / (FP + TP)\n",
"# extra code – this cell also computes the precision: TP / (FP + TP)\n",
"cm[1, 1] / (cm[0, 1] + cm[1, 1])"
]
},
@ -446,7 +446,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code also computes the recall: TP / (FN + TP)\n",
"# extra code – this cell also computes the recall: TP / (FN + TP)\n",
"cm[1, 1] / (cm[1, 0] + cm[1, 1])"
]
},
@ -467,7 +467,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code also computes the f1 score\n",
"# extra code – this cell also computes the f1 score\n",
"cm[1, 1] / (cm[1, 1] + (cm[1, 0] + cm[0, 1]) / 2)"
]
},
@ -513,8 +513,8 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code just shows that y_scores > 0 produces the same\n",
"# result as calling predict()\n",
"# extra code – just shows that y_scores > 0 produces the same result as\n",
"# calling predict()\n",
"y_scores > 0"
]
},
@ -556,12 +556,12 @@
"metadata": {},
"outputs": [],
"source": [
"plt.figure(figsize=(8, 4)) # not in the book – it's not needed, just formatting\n",
"plt.figure(figsize=(8, 4)) # extra code – it's not needed, just formatting\n",
"plt.plot(thresholds, precisions[:-1], \"b--\", label=\"Precision\", linewidth=2)\n",
"plt.plot(thresholds, recalls[:-1], \"g-\", label=\"Recall\", linewidth=2)\n",
"plt.vlines(threshold, 0, 1.0, \"k\", \"dotted\", label=\"threshold\")\n",
"\n",
"# not in the book – this section just beautifies and saves Figure 3–5\n",
"# extra code – this section just beautifies and saves Figure 3–5\n",
"idx = (thresholds >= threshold).argmax() # first index ≥ threshold\n",
"plt.plot(thresholds[idx], precisions[idx], \"bo\")\n",
"plt.plot(thresholds[idx], recalls[idx], \"go\")\n",
@ -580,13 +580,13 @@
"metadata": {},
"outputs": [],
"source": [
"import matplotlib.patches as patches # not in the book – for the curved arrow\n",
"import matplotlib.patches as patches # extra code – for the curved arrow\n",
"\n",
"plt.figure(figsize=(6, 5)) # not in the book – not needed, just formatting\n",
"plt.figure(figsize=(6, 5)) # extra code – not needed, just formatting\n",
"\n",
"plt.plot(recalls, precisions, linewidth=2, label=\"Precision/Recall curve\")\n",
"\n",
"# not in the book – just beautifies and saves Figure 3–6\n",
"# extra code – just beautifies and saves Figure 3–6\n",
"plt.plot([recalls[idx], recalls[idx]], [0., precisions[idx]], \"k:\")\n",
"plt.plot([0.0, recalls[idx]], [precisions[idx], precisions[idx]], \"k:\")\n",
"plt.plot([recalls[idx]], [precisions[idx]], \"ko\",\n",
@ -673,12 +673,12 @@
"idx_for_threshold_at_90 = (thresholds <= threshold_for_90_precision).argmax()\n",
"tpr_90, fpr_90 = tpr[idx_for_threshold_at_90], fpr[idx_for_threshold_at_90]\n",
"\n",
"plt.figure(figsize=(6, 5)) # not in the book – not needed, just formatting\n",
"plt.figure(figsize=(6, 5)) # extra code – not needed, just formatting\n",
"plt.plot(fpr, tpr, linewidth=2, label=\"ROC curve\")\n",
"plt.plot([0, 1], [0, 1], 'k:', label=\"Random classifier's ROC curve\")\n",
"plt.plot([fpr_90], [tpr_90], \"ko\", label=\"Threshold for 90% precision\")\n",
"\n",
"# not in the book – just beautifies and saves Figure 3–7\n",
"# extra code – just beautifies and saves Figure 3–7\n",
"plt.gca().add_patch(patches.FancyArrowPatch(\n",
" (0.20, 0.89), (0.07, 0.70),\n",
" connectionstyle=\"arc3,rad=.4\",\n",
@ -778,13 +778,13 @@
"metadata": {},
"outputs": [],
"source": [
"plt.figure(figsize=(6, 5)) # not in the book – not needed, just formatting\n",
"plt.figure(figsize=(6, 5)) # extra code – not needed, just formatting\n",
"\n",
"plt.plot(recalls_forest, precisions_forest, \"b-\", linewidth=2,\n",
" label=\"Random Forest\")\n",
"plt.plot(recalls, precisions, \"--\", linewidth=2, label=\"SGD\")\n",
"\n",
"# not in the book – just beautifies and saves Figure 3–8\n",
"# extra code – just beautifies and saves Figure 3–8\n",
"plt.xlabel(\"Recall\")\n",
"plt.ylabel(\"Precision\")\n",
"plt.axis([0, 1, 0, 1])\n",
@ -925,7 +925,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code shows how to get all 45 OvO scores if needed\n",
"# extra code – shows how to get all 45 OvO scores if needed\n",
"svm_clf.decision_function_shape = \"ovo\"\n",
"some_digit_scores_ovo = svm_clf.decision_function([some_digit])\n",
"some_digit_scores_ovo.round(2)"
@ -1033,7 +1033,7 @@
"from sklearn.metrics import ConfusionMatrixDisplay\n",
"\n",
"y_train_pred = cross_val_predict(sgd_clf, X_train_scaled, y_train, cv=3)\n",
"plt.rc('font', size=9) # not in the book – make the text smaller\n",
"plt.rc('font', size=9) # extra code – make the text smaller\n",
"ConfusionMatrixDisplay.from_predictions(y_train, y_train_pred)\n",
"plt.show()"
]
@ -1044,7 +1044,7 @@
"metadata": {},
"outputs": [],
"source": [
"plt.rc('font', size=10) # not in the book\n",
"plt.rc('font', size=10) # extra code\n",
"ConfusionMatrixDisplay.from_predictions(y_train, y_train_pred,\n",
" normalize=\"true\", values_format=\".0%\")\n",
"plt.show()"
@ -1057,7 +1057,7 @@
"outputs": [],
"source": [
"sample_weight = (y_train_pred != y_train)\n",
"plt.rc('font', size=10) # not in the book\n",
"plt.rc('font', size=10) # extra code\n",
"ConfusionMatrixDisplay.from_predictions(y_train, y_train_pred,\n",
" sample_weight=sample_weight,\n",
" normalize=\"true\", values_format=\".0%\")\n",
@ -1077,7 +1077,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates Figure 3–9\n",
"# extra code – this cell generates and saves Figure 3–9\n",
"fig, axs = plt.subplots(nrows=1, ncols=2, figsize=(9, 4))\n",
"plt.rc('font', size=9)\n",
"ConfusionMatrixDisplay.from_predictions(y_train, y_train_pred, ax=axs[0])\n",
@ -1096,7 +1096,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates Figure 3–10\n",
"# extra code – this cell generates and saves Figure 3–10\n",
"fig, axs = plt.subplots(nrows=1, ncols=2, figsize=(9, 4))\n",
"plt.rc('font', size=10)\n",
"ConfusionMatrixDisplay.from_predictions(y_train, y_train_pred, ax=axs[0],\n",
@ -1131,7 +1131,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates Figure 3–11\n",
"# extra code – this cell generates and saves Figure 3–11\n",
"size = 5\n",
"pad = 0.2\n",
"plt.figure(figsize=(size, size))\n",
@ -1224,9 +1224,9 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code shows that we get a negligible performance\n",
"# improvement when we set average=\"weighted\" because the\n",
"# classes are already pretty well balanced.\n",
"# extra code – shows that we get a negligible performance improvement when we\n",
"# set average=\"weighted\" because the classes are already pretty\n",
"# well balanced.\n",
"f1_score(y_multilabel, y_train_knn_pred, average=\"weighted\")"
]
},
@ -1279,7 +1279,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates Figure 3–12\n",
"# extra code – this cell generates and saves Figure 3–12\n",
"plt.subplot(121); plot_digit(X_test_mod[0])\n",
"plt.subplot(122); plot_digit(y_test_mod[0])\n",
"save_fig(\"noisy_digit_example_plot\")\n",
@ -1296,7 +1296,7 @@
"knn_clf.fit(X_train_mod, y_train_mod)\n",
"clean_digit = knn_clf.predict([X_test_mod[0]])\n",
"plot_digit(clean_digit)\n",
"save_fig(\"cleaned_digit_example_plot\") # not in the book – saves Figure 3–13\n",
"save_fig(\"cleaned_digit_example_plot\") # extra code – saves Figure 3–13\n",
"plt.show()"
]
},

235
04_training_linear_models.ipynb
View File

@ -91,8 +91,8 @@
"plt.rc('font', size=14)\n",
"plt.rc('axes', labelsize=14, titlesize=14)\n",
"plt.rc('legend', fontsize=14)\n",
"plt.rc('xtick',labelsize=10)\n",
"plt.rc('ytick',labelsize=10)"
"plt.rc('xtick', labelsize=10)\n",
"plt.rc('ytick', labelsize=10)"
]
},
{
@ -154,7 +154,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – generates and saves Figure 4–1\n",
"# extra code – generates and saves Figure 4–1\n",
"\n",
"import matplotlib.pyplot as plt\n",
"\n",
@ -209,11 +209,11 @@
"source": [
"import matplotlib.pyplot as plt\n",
"\n",
"plt.figure(figsize=(6, 4)) # not in the book – not needed, just formatting\n",
"plt.figure(figsize=(6, 4)) # extra code – not needed, just formatting\n",
"plt.plot(X_new, y_predict, \"r-\", label=\"Predictions\")\n",
"plt.plot(X, y, \"b.\")\n",
"\n",
"# not in the book – beautifies and saves Figure 4–2\n",
"# extra code – beautifies and saves Figure 4–2\n",
"plt.xlabel(\"$x_1$\")\n",
"plt.ylabel(\"$y$\", rotation=0)\n",
"plt.axis([0, 2, 0, 15])\n",
@ -327,7 +327,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – generates and saves Figure 4–8\n",
"# extra code – generates and saves Figure 4–8\n",
"\n",
"import matplotlib as mpl\n",
"\n",
@ -352,9 +352,9 @@
" return theta_path\n",
"\n",
"np.random.seed(42)\n",
"theta = np.random.randn(2,1) # random initialization\n",
"theta = np.random.randn(2, 1) # random initialization\n",
"\n",
"plt.figure(figsize=(10,4))\n",
"plt.figure(figsize=(10, 4))\n",
"plt.subplot(131)\n",
"plot_gradient_descent(theta, eta=0.02)\n",
"plt.ylabel(\"$y$\", rotation=0)\n",
@ -381,8 +381,8 @@
"metadata": {},
"outputs": [],
"source": [
"theta_path_sgd = [] # not in the book – we need to store the path of theta in\n",
" # the parameter space to plot the next figure"
"theta_path_sgd = [] # extra code – we need to store the path of theta in the\n",
" # parameter space to plot the next figure"
]
},
{
@ -400,13 +400,13 @@
"np.random.seed(42)\n",
"theta = np.random.randn(2, 1) # random initialization\n",
"\n",
"n_shown = 20 # not in the book – just needed to generate the figure below\n",
"plt.figure(figsize=(6, 4)) # not in the book – not needed, just formatting\n",
"n_shown = 20 # extra code – just needed to generate the figure below\n",
"plt.figure(figsize=(6, 4)) # extra code – not needed, just formatting\n",
"\n",
"for epoch in range(n_epochs):\n",
" for iteration in range(m):\n",
"\n",
" # not in the book – these 4 lines are used to generate the figure\n",
" # extra code – these 4 lines are used to generate the figure\n",
" if epoch == 0 and iteration < n_shown:\n",
" y_predict = X_new_b @ theta\n",
" color = mpl.colors.rgb2hex(plt.cm.OrRd(iteration / n_shown + 0.15))\n",
@ -415,12 +415,12 @@
" random_index = np.random.randint(m)\n",
" xi = X_b[random_index : random_index + 1]\n",
" yi = y[random_index : random_index + 1]\n",
" gradients = 2 / 1 * xi.T @ (xi @ theta - yi)\n",
" gradients = 2 * xi.T @ (xi @ theta - yi) # for SGD, do not divide by m\n",
" eta = learning_schedule(epoch * m + iteration)\n",
" theta = theta - eta * gradients\n",
" theta_path_sgd.append(theta) # not in the book – to generate the figure\n",
" theta_path_sgd.append(theta) # extra code – to generate the figure\n",
"\n",
"# not in the book – this section beautifies and saves Figure 4–10\n",
"# extra code – this section beautifies and saves Figure 4–10\n",
"plt.plot(X, y, \"b.\")\n",
"plt.xlabel(\"$x_1$\")\n",
"plt.ylabel(\"$y$\", rotation=0)\n",
@ -449,9 +449,9 @@
"source": [
"from sklearn.linear_model import SGDRegressor\n",
"\n",
"sgd_reg = SGDRegressor(max_iter=1000, tol=1e-3, penalty=None, eta0=0.1,\n",
" random_state=42)\n",
"sgd_reg.fit(X, y.ravel()) # y.ravel() because fit() expects 1D targets"
"sgd_reg = SGDRegressor(max_iter=1000, tol=1e-5, penalty=None, eta0=0.01,\n",
" n_iter_no_change=100, random_state=42)\n",
"sgd_reg.fit(X, y.ravel()) # y.ravel() because fit() expects 1D targets\n"
]
},
{
@ -483,7 +483,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 4–11\n",
"# extra code – this cell generates and saves Figure 4–11\n",
"\n",
"from math import ceil\n",
"\n",
@ -558,7 +558,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 4–12\n",
"# extra code – this cell generates and saves Figure 4–12\n",
"plt.figure(figsize=(6, 4))\n",
"plt.plot(X, y, \"b.\")\n",
"plt.xlabel(\"$x_1$\")\n",
@ -608,7 +608,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 4–13\n",
"# extra code – this cell generates and saves Figure 4–13\n",
"\n",
"X_new = np.linspace(-3, 3, 100).reshape(100, 1)\n",
"X_new_poly = poly_features.transform(X_new)\n",
@ -632,7 +632,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 4–14\n",
"# extra code – this cell generates and saves Figure 4–14\n",
"\n",
"from sklearn.preprocessing import StandardScaler\n",
"from sklearn.pipeline import make_pipeline\n",
@ -680,11 +680,11 @@
"train_errors = -train_scores.mean(axis=1)\n",
"valid_errors = -valid_scores.mean(axis=1)\n",
"\n",
"plt.figure(figsize=(6, 4)) # not in the book – not need, just formatting\n",
"plt.figure(figsize=(6, 4)) # extra code – not need, just formatting\n",
"plt.plot(train_sizes, train_errors, \"r-+\", linewidth=2, label=\"train\")\n",
"plt.plot(train_sizes, valid_errors, \"b-\", linewidth=3, label=\"valid\")\n",
"\n",
"# not in the book – beautifies and saves Figure 4–15\n",
"# extra code – beautifies and saves Figure 4–15\n",
"plt.xlabel(\"Training set size\")\n",
"plt.ylabel(\"RMSE\")\n",
"plt.grid()\n",
@ -718,7 +718,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – generates and saves Figure 4–16\n",
"# extra code – generates and saves Figure 4–16\n",
"\n",
"train_errors = -train_scores.mean(axis=1)\n",
"valid_errors = -valid_scores.mean(axis=1)\n",
@ -762,7 +762,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – we've done this type of generation several times before\n",
"# extra code – we've done this type of generation several times before\n",
"np.random.seed(42)\n",
"m = 20\n",
"X = 3 * np.random.rand(m, 1)\n",
@ -776,7 +776,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – a quick peek at the dataset we just generated\n",
"# extra code – a quick peek at the dataset we just generated\n",
"plt.figure(figsize=(6, 4))\n",
"plt.plot(X, y, \".\")\n",
"plt.xlabel(\"$x_1$\")\n",
@ -794,7 +794,7 @@
"source": [
"from sklearn.linear_model import Ridge\n",
"\n",
"ridge_reg = Ridge(alpha=1, solver=\"cholesky\")\n",
"ridge_reg = Ridge(alpha=0.1, solver=\"cholesky\")\n",
"ridge_reg.fit(X, y)\n",
"ridge_reg.predict([[1.5]])"
]
@ -805,7 +805,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 4–17\n",
"# extra code – this cell generates and saves Figure 4–17\n",
"\n",
"def plot_model(model_class, polynomial, alphas, **model_kargs):\n",
" plt.plot(X, y, \"b.\", linewidth=3)\n",
@ -845,8 +845,9 @@
"metadata": {},
"outputs": [],
"source": [
"sgd_reg = SGDRegressor(penalty=\"l2\", random_state=42)\n",
"sgd_reg.fit(X, y.ravel())\n",
"sgd_reg = SGDRegressor(penalty=\"l2\", alpha=0.1 / m, tol=None,\n",
" max_iter=1000, eta0=0.01, random_state=42)\n",
"sgd_reg.fit(X, y.ravel()) # y.ravel() because fit() expects 1D targets\n",
"sgd_reg.predict([[1.5]])"
]
},
@ -856,13 +857,36 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – show that we get roughly the same solution as earlier when\n",
"# we use Stochastic Average GD (solver=\"sag\")\n",
"ridge_reg = Ridge(alpha=1, solver=\"sag\", random_state=42)\n",
"# extra code – show that we get roughly the same solution as earlier when\n",
"# we use Stochastic Average GD (solver=\"sag\")\n",
"ridge_reg = Ridge(alpha=0.1, solver=\"sag\", random_state=42)\n",
"ridge_reg.fit(X, y)\n",
"ridge_reg.predict([[1.5]])"
]
},
{
"cell_type": "code",
"execution_count": 40,
"metadata": {},
"outputs": [],
"source": [
"# extra code – shows the closed form solution of Ridge regression,\n",
"# compare with the next Ridge model's learned parameters below\n",
"alpha = 0.1\n",
"A = np.array([[0., 0.], [0., 1.]])\n",
"X_b = np.c_[np.ones(m), X]\n",
"np.linalg.inv(X_b.T @ X_b + alpha * A) @ X_b.T @ y"
]
},
{
"cell_type": "code",
"execution_count": 41,
"metadata": {},
"outputs": [],
"source": [
"ridge_reg.intercept_, ridge_reg.coef_ # extra code"
]
},
{
"cell_type": "markdown",
"metadata": {},
@ -872,7 +896,7 @@
},
{
"cell_type": "code",
"execution_count": 40,
"execution_count": 42,
"metadata": {},
"outputs": [],
"source": [
@ -885,11 +909,11 @@
},
{
"cell_type": "code",
"execution_count": 41,
"execution_count": 43,
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 4–18\n",
"# extra code – this cell generates and saves Figure 4–18\n",
"plt.figure(figsize=(9, 3.5))\n",
"plt.subplot(121)\n",
"plot_model(Lasso, polynomial=False, alphas=(0, 0.1, 1), random_state=42)\n",
@ -903,11 +927,11 @@
},
{
"cell_type": "code",
"execution_count": 42,
"execution_count": 44,
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this BIG cell generates and saves Figure 4–19\n",
"# extra code – this BIG cell generates and saves Figure 4–19\n",
"\n",
"t1a, t1b, t2a, t2b = -1, 3, -1.5, 1.5\n",
"\n",
@ -996,7 +1020,7 @@
},
{
"cell_type": "code",
"execution_count": 43,
"execution_count": 45,
"metadata": {},
"outputs": [],
"source": [
@ -1023,22 +1047,7 @@
},
{
"cell_type": "code",
"execution_count": 44,
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this is the same code as earlier\n",
"np.random.seed(42)\n",
"m = 100\n",
"X = 6 * np.random.rand(m, 1) - 3\n",
"y = 0.5 * X ** 2 + X + 2 + np.random.randn(m, 1)\n",
"X_train, y_train = X[: m // 2], y[: m // 2, 0]\n",
"X_valid, y_valid = X[m // 2 :], y[m // 2 :, 0]"
]
},
{
"cell_type": "code",
"execution_count": 45,
"execution_count": 46,
"metadata": {},
"outputs": [],
"source": [
@ -1046,6 +1055,14 @@
"from sklearn.metrics import mean_squared_error\n",
"from sklearn.preprocessing import StandardScaler\n",
"\n",
"# extra code – creates the same quadratic dataset as earlier and splits it\n",
"np.random.seed(42)\n",
"m = 100\n",
"X = 6 * np.random.rand(m, 1) - 3\n",
"y = 0.5 * X ** 2 + X + 2 + np.random.randn(m, 1)\n",
"X_train, y_train = X[: m // 2], y[: m // 2, 0]\n",
"X_valid, y_valid = X[m // 2 :], y[m // 2 :, 0]\n",
"\n",
"preprocessing = make_pipeline(PolynomialFeatures(degree=90, include_bias=False),\n",
" StandardScaler())\n",
"X_train_prep = preprocessing.fit_transform(X_train)\n",
@ -1053,7 +1070,7 @@
"sgd_reg = SGDRegressor(penalty=None, eta0=0.002, random_state=42)\n",
"n_epochs = 500\n",
"best_valid_rmse = float('inf')\n",
"train_errors, val_errors = [], [] # not in the book – it's for the figure below\n",
"train_errors, val_errors = [], [] # extra code – it's for the figure below\n",
"\n",
"for epoch in range(n_epochs):\n",
" sgd_reg.partial_fit(X_train_prep, y_train)\n",
@ -1063,13 +1080,13 @@
" best_valid_rmse = val_error\n",
" best_model = deepcopy(sgd_reg)\n",
"\n",
" # not in the book – we evaluate the train error and save it for the figure\n",
" # extra code – we evaluate the train error and save it for the figure\n",
" y_train_predict = sgd_reg.predict(X_train_prep)\n",
" train_error = mean_squared_error(y_train, y_train_predict, squared=False)\n",
" val_errors.append(val_error)\n",
" train_errors.append(train_error)\n",
"\n",
"# not in the book – this section generates and saves Figure 4–20\n",
"# extra code – this section generates and saves Figure 4–20\n",
"best_epoch = np.argmin(val_errors)\n",
"plt.figure(figsize=(6, 4))\n",
"plt.annotate('Best model',\n",
@ -1106,11 +1123,11 @@
},
{
"cell_type": "code",
"execution_count": 46,
"execution_count": 47,
"metadata": {},
"outputs": [],
"source": [
"# not in the book – generates and saves Figure 4–21\n",
"# extra code – generates and saves Figure 4–21\n",
"\n",
"lim = 6\n",
"t = np.linspace(-lim, lim, 100)\n",
@ -1140,7 +1157,7 @@
},
{
"cell_type": "code",
"execution_count": 47,
"execution_count": 48,
"metadata": {},
"outputs": [],
"source": [
@ -1150,22 +1167,13 @@
"list(iris)"
]
},
{
"cell_type": "code",
"execution_count": 48,
"metadata": {},
"outputs": [],
"source": [
"print(iris.DESCR) # not in the book – it's a bit too long"
]
},
{
"cell_type": "code",
"execution_count": 49,
"metadata": {},
"outputs": [],
"source": [
"iris.data.head(3)"
"print(iris.DESCR) # extra code – it's a bit too long"
]
},
{
@ -1174,7 +1182,7 @@
"metadata": {},
"outputs": [],
"source": [
"iris.target.head(3) # note that the instances are not shuffled"
"iris.data.head(3)"
]
},
{
@ -1183,7 +1191,7 @@
"metadata": {},
"outputs": [],
"source": [
"iris.target_names"
"iris.target.head(3) # note that the instances are not shuffled"
]
},
{
@ -1191,6 +1199,15 @@
"execution_count": 52,
"metadata": {},
"outputs": [],
"source": [
"iris.target_names"
]
},
{
"cell_type": "code",
"execution_count": 53,
"metadata": {},
"outputs": [],
"source": [
"from sklearn.linear_model import LogisticRegression\n",
"from sklearn.model_selection import train_test_split\n",
@ -1205,7 +1222,7 @@
},
{
"cell_type": "code",
"execution_count": 53,
"execution_count": 54,
"metadata": {},
"outputs": [],
"source": [
@ -1213,14 +1230,14 @@
"y_proba = log_reg.predict_proba(X_new)\n",
"decision_boundary = X_new[y_proba[:, 1] >= 0.5][0, 0]\n",
"\n",
"plt.figure(figsize=(8, 3)) # not in the book – not needed, just formatting\n",
"plt.figure(figsize=(8, 3)) # extra code – not needed, just formatting\n",
"plt.plot(X_new, y_proba[:, 0], \"b--\", linewidth=2,\n",
" label=\"Not Iris virginica proba\")\n",
"plt.plot(X_new, y_proba[:, 1], \"g-\", linewidth=2, label=\"Iris virginica proba\")\n",
"plt.plot([decision_boundary, decision_boundary], [0, 1], \"k:\", linewidth=2,\n",
" label=\"Decision boundary\")\n",
"\n",
"# not in the book – this section beautifies and saves Figure 4–21\n",
"# extra code – this section beautifies and saves Figure 4–21\n",
"plt.arrow(x=decision_boundary, y=0.08, dx=-0.3, dy=0,\n",
" head_width=0.05, head_length=0.1, fc=\"b\", ec=\"b\")\n",
"plt.arrow(x=decision_boundary, y=0.92, dx=0.3, dy=0,\n",
@ -1239,7 +1256,7 @@
},
{
"cell_type": "code",
"execution_count": 54,
"execution_count": 55,
"metadata": {},
"outputs": [],
"source": [
@ -1248,7 +1265,7 @@
},
{
"cell_type": "code",
"execution_count": 55,
"execution_count": 56,
"metadata": {},
"outputs": [],
"source": [
@ -1257,11 +1274,11 @@
},
{
"cell_type": "code",
"execution_count": 56,
"execution_count": 57,
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 4–22\n",
"# extra code – this cell generates and saves Figure 4–22\n",
"\n",
"X = iris.data[[\"petal length (cm)\", \"petal width (cm)\"]].values\n",
"y = iris.target_names[iris.target] == 'virginica'\n",
@ -1307,7 +1324,7 @@
},
{
"cell_type": "code",
"execution_count": 57,
"execution_count": 58,
"metadata": {},
"outputs": [],
"source": [
@ -1321,7 +1338,7 @@
},
{
"cell_type": "code",
"execution_count": 58,
"execution_count": 59,
"metadata": {
"tags": []
},
@ -1332,7 +1349,7 @@
},
{
"cell_type": "code",
"execution_count": 59,
"execution_count": 60,
"metadata": {
"tags": []
},
@ -1343,11 +1360,11 @@
},
{
"cell_type": "code",
"execution_count": 60,
"execution_count": 61,
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 4–23\n",
"# extra code – this cell generates and saves Figure 4–23\n",
"\n",
"from matplotlib.colors import ListedColormap\n",
"\n",
@ -1419,7 +1436,7 @@
"metadata": {},
"source": [
"## 12. Batch Gradient Descent with early stopping for Softmax Regression\n",
"Exercise: _Implement Batch Gradient Descent with early stopping for Softmax Regression without using Scikit-Learn, only NumPy._"
"Exercise: _Implement Batch Gradient Descent with early stopping for Softmax Regression without using Scikit-Learn, only NumPy. Use it on a classification task such as the iris dataset._"
]
},
{
@ -1431,7 +1448,7 @@
},
{
"cell_type": "code",
"execution_count": 61,
"execution_count": 62,
"metadata": {},
"outputs": [],
"source": [
@ -1448,7 +1465,7 @@
},
{
"cell_type": "code",
"execution_count": 62,
"execution_count": 63,
"metadata": {},
"outputs": [],
"source": [
@ -1464,7 +1481,7 @@
},
{
"cell_type": "code",
"execution_count": 63,
"execution_count": 64,
"metadata": {},
"outputs": [],
"source": [
@ -1491,12 +1508,12 @@
"cell_type": "markdown",
"metadata": {},
"source": [
"The targets are currently class indices (0, 1 or 2), but we need target class probabilities to train the Softmax Regression model. Each instance will have target class probabilities equal to 0.0 for all classes except for the target class which will have a probability of 1.0 (in other words, the vector of class probabilities for any given instance is a one-hot vector). Let's write a small function to convert the vector of class indices into a matrix containing a one-hot vector for each instance. To understand this code, you need to know that `np.diag(np.ones(n))` creates an n×n matrix full of 0s except for 1s on the main diagonal. Moreover, if `a` in a NumPy array, then `a[[1,3,2]]` returns an array with 3 rows equal to `a[1]`, `a[3]` and `a[2]` (this is [advanced NumPy indexing](https://numpy.org/doc/stable/reference/arrays.indexing.html#advanced-indexing))."
"The targets are currently class indices (0, 1 or 2), but we need target class probabilities to train the Softmax Regression model. Each instance will have target class probabilities equal to 0.0 for all classes except for the target class which will have a probability of 1.0 (in other words, the vector of class probabilities for any given instance is a one-hot vector). Let's write a small function to convert the vector of class indices into a matrix containing a one-hot vector for each instance. To understand this code, you need to know that `np.diag(np.ones(n))` creates an n×n matrix full of 0s except for 1s on the main diagonal. Moreover, if `a` in a NumPy array, then `a[[1, 3, 2]]` returns an array with 3 rows equal to `a[1]`, `a[3]` and `a[2]` (this is [advanced NumPy indexing](https://numpy.org/doc/stable/reference/arrays.indexing.html#advanced-indexing))."
]
},
{
"cell_type": "code",
"execution_count": 64,
"execution_count": 65,
"metadata": {},
"outputs": [],
"source": [
@ -1513,7 +1530,7 @@
},
{
"cell_type": "code",
"execution_count": 65,
"execution_count": 66,
"metadata": {},
"outputs": [],
"source": [
@ -1522,7 +1539,7 @@
},
{
"cell_type": "code",
"execution_count": 66,
"execution_count": 67,
"metadata": {},
"outputs": [],
"source": [
@ -1538,7 +1555,7 @@
},
{
"cell_type": "code",
"execution_count": 67,
"execution_count": 68,
"metadata": {},
"outputs": [],
"source": [
@ -1556,7 +1573,7 @@
},
{
"cell_type": "code",
"execution_count": 68,
"execution_count": 69,
"metadata": {},
"outputs": [],
"source": [
@ -1578,7 +1595,7 @@
},
{
"cell_type": "code",
"execution_count": 69,
"execution_count": 70,
"metadata": {},
"outputs": [],
"source": [
@ -1597,7 +1614,7 @@
},
{
"cell_type": "code",
"execution_count": 70,
"execution_count": 71,
"metadata": {},
"outputs": [],
"source": [
@ -1625,7 +1642,7 @@
},
{
"cell_type": "code",
"execution_count": 71,
"execution_count": 72,
"metadata": {},
"outputs": [],
"source": [
@ -1658,7 +1675,7 @@
},
{
"cell_type": "code",
"execution_count": 72,
"execution_count": 73,
"metadata": {},
"outputs": [],
"source": [
@ -1674,7 +1691,7 @@
},
{
"cell_type": "code",
"execution_count": 73,
"execution_count": 74,
"metadata": {},
"outputs": [],
"source": [
@ -1695,7 +1712,7 @@
},
{
"cell_type": "code",
"execution_count": 74,
"execution_count": 75,
"metadata": {},
"outputs": [],
"source": [
@ -1732,7 +1749,7 @@
},
{
"cell_type": "code",
"execution_count": 75,
"execution_count": 76,
"metadata": {},
"outputs": [],
"source": [
@ -1760,7 +1777,7 @@
},
{
"cell_type": "code",
"execution_count": 76,
"execution_count": 77,
"metadata": {},
"outputs": [],
"source": [
@ -1797,7 +1814,7 @@
},
{
"cell_type": "code",
"execution_count": 77,
"execution_count": 78,
"metadata": {},
"outputs": [],
"source": [
@ -1825,11 +1842,11 @@
},
{
"cell_type": "code",
"execution_count": 78,
"execution_count": 79,
"metadata": {},
"outputs": [],
"source": [
"custom_cmap = mpl.colors.ListedColormap(['#fafab0','#9898ff','#a0faa0'])\n",
"custom_cmap = mpl.colors.ListedColormap(['#fafab0', '#9898ff', '#a0faa0'])\n",
"\n",
"x0, x1 = np.meshgrid(np.linspace(0, 8, 500).reshape(-1, 1),\n",
" np.linspace(0, 3.5, 200).reshape(-1, 1))\n",
@ -1869,7 +1886,7 @@
},
{
"cell_type": "code",
"execution_count": 79,
"execution_count": 80,
"metadata": {},
"outputs": [],
"source": [

32
05_support_vector_machines.ipynb
View File

@ -140,7 +140,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 5–1\n",
"# extra code – this cell generates and saves Figure 5–1\n",
"\n",
"import matplotlib.pyplot as plt\n",
"import numpy as np\n",
@ -219,7 +219,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 5–2\n",
"# extra code – this cell generates and saves Figure 5–2\n",
"\n",
"from sklearn.preprocessing import StandardScaler\n",
"\n",
@ -269,7 +269,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 5–3\n",
"# extra code – this cell generates and saves Figure 5–3\n",
"\n",
"X_outliers = np.array([[3.4, 1.3], [3.2, 0.8]])\n",
"y_outliers = np.array([0, 0])\n",
@ -364,7 +364,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 5–4\n",
"# extra code – this cell generates and saves Figure 5–4\n",
"\n",
"scaler = StandardScaler()\n",
"svm_clf1 = LinearSVC(C=1, max_iter=10_000, random_state=42)\n",
@ -432,7 +432,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 5–5\n",
"# extra code – this cell generates and saves Figure 5–5\n",
"\n",
"X1D = np.linspace(-4, 4, 9).reshape(-1, 1)\n",
"X2D = np.c_[X1D, X1D**2]\n",
@ -492,7 +492,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 5–6\n",
"# extra code – this cell generates and saves Figure 5–6\n",
"\n",
"def plot_dataset(X, y, axes):\n",
" plt.plot(X[:, 0][y==0], X[:, 1][y==0], \"bs\")\n",
@ -545,7 +545,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 5–7\n",
"# extra code – this cell generates and saves Figure 5–7\n",
"\n",
"poly100_kernel_svm_clf = make_pipeline(\n",
" StandardScaler(),\n",
@ -585,7 +585,7 @@
},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 5–8\n",
"# extra code – this cell generates and saves Figure 5–8\n",
"\n",
"def gaussian_rbf(x, landmark, gamma):\n",
" return np.exp(-gamma * np.linalg.norm(x - landmark, axis=1)**2)\n",
@ -675,7 +675,7 @@
},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 5–9\n",
"# extra code – this cell generates and saves Figure 5–9\n",
"\n",
"from sklearn.svm import SVC\n",
"\n",
@ -724,7 +724,7 @@
"source": [
"from sklearn.svm import LinearSVR\n",
"\n",
"# not in the book – these 3 lines generate a simple linear dataset\n",
"# extra code – these 3 lines generate a simple linear dataset\n",
"np.random.seed(42)\n",
"X = 2 * np.random.rand(50, 1)\n",
"y = 4 + 3 * X[:, 0] + np.random.randn(50)\n",
@ -740,7 +740,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 5–10\n",
"# extra code – this cell generates and saves Figure 5–10\n",
"\n",
"def find_support_vectors(svm_reg, X, y):\n",
" y_pred = svm_reg.predict(X)\n",
@ -800,7 +800,7 @@
"source": [
"from sklearn.svm import SVR\n",
"\n",
"# not in the book – these 3 lines generate a simple quadratic dataset\n",
"# extra code – these 3 lines generate a simple quadratic dataset\n",
"np.random.seed(42)\n",
"X = 2 * np.random.rand(50, 1) - 1\n",
"y = 0.2 + 0.1 * X[:, 0] + 0.5 * X[:, 0] ** 2 + np.random.randn(50) / 10\n",
@ -816,7 +816,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 5–11\n",
"# extra code – this cell generates and saves Figure 5–11\n",
"\n",
"svm_poly_reg2 = make_pipeline(StandardScaler(),\n",
" SVR(kernel=\"poly\", degree=2, C=100))\n",
@ -857,7 +857,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 5–12\n",
"# extra code – this cell generates and saves Figure 5–12\n",
"\n",
"import matplotlib.patches as patches\n",
"\n",
@ -906,7 +906,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 5–13\n",
"# extra code – this cell generates and saves Figure 5–13\n",
"\n",
"s = np.linspace(-2.5, 2.5, 200)\n",
"hinge_pos = np.where(1 - s < 0, 0, 1 - s) # max(0, 1 - s)\n",
@ -1112,7 +1112,7 @@
"sgd_clf.fit(X, y)\n",
"\n",
"m = len(X)\n",
"t = np.array(y).reshape(-1, 1) * 2 - 1 # -1 if t==0, +1 if t==1\n",
"t = np.array(y).reshape(-1, 1) * 2 - 1 # -1 if y == 0, or +1 if y == 1\n",
"X_b = np.c_[np.ones((m, 1)), X] # Add bias input x0=1\n",
"X_b_t = X_b * t\n",
"sgd_theta = np.r_[sgd_clf.intercept_[0], sgd_clf.coef_[0]]\n",

20
06_decision_trees.ipynb
View File

@ -193,7 +193,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book\n",
"# extra code\n",
"!dot -Tpng {IMAGES_PATH / \"iris_tree.dot\"} -o {IMAGES_PATH / \"iris_tree.png\"}"
]
},
@ -213,7 +213,7 @@
"import numpy as np\n",
"import matplotlib.pyplot as plt\n",
"\n",
"# not in the book – just formatting details\n",
"# extra code – just formatting details\n",
"from matplotlib.colors import ListedColormap\n",
"custom_cmap = ListedColormap(['#fafab0', '#9898ff', '#a0faa0'])\n",
"plt.figure(figsize=(8, 4))\n",
@ -226,7 +226,7 @@
" plt.plot(X_iris[:, 0][y_iris == idx], X_iris[:, 1][y_iris == idx],\n",
" style, label=f\"Iris {name}\")\n",
"\n",
"# not in the book – this section beautifies and saves Figure 6–2\n",
"# extra code – this section beautifies and saves Figure 6–2\n",
"tree_clf_deeper = DecisionTreeClassifier(max_depth=3, random_state=42)\n",
"tree_clf_deeper.fit(X_iris, y_iris)\n",
"th0, th1, th2a, th2b = tree_clf_deeper.tree_.threshold[[0, 2, 3, 6]]\n",
@ -341,7 +341,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 6–3\n",
"# extra code – this cell generates and saves Figure 6–3\n",
"\n",
"def plot_decision_boundary(clf, X, y, axes, cmap):\n",
" x1, x2 = np.meshgrid(np.linspace(axes[0], axes[1], 100),\n",
@ -437,7 +437,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – we've already seen how to use export_graphviz()\n",
"# extra code – we've already seen how to use export_graphviz()\n",
"export_graphviz(\n",
" tree_reg,\n",
" out_file=str(IMAGES_PATH / \"regression_tree.dot\"),\n",
@ -482,7 +482,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 6–5\n",
"# extra code – this cell generates and saves Figure 6–5\n",
"\n",
"def plot_regression_predictions(tree_reg, X, y, axes=[-0.5, 0.5, -0.05, 0.25]):\n",
" x1 = np.linspace(axes[0], axes[1], 500).reshape(-1, 1)\n",
@ -526,7 +526,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 6–6\n",
"# extra code – this cell generates and saves Figure 6–6\n",
"\n",
"tree_reg1 = DecisionTreeRegressor(random_state=42)\n",
"tree_reg2 = DecisionTreeRegressor(random_state=42, min_samples_leaf=10)\n",
@ -579,7 +579,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 6–7\n",
"# extra code – this cell generates and saves Figure 6–7\n",
"\n",
"np.random.seed(6)\n",
"X_square = np.random.rand(100, 2) - 0.5\n",
@ -630,7 +630,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 6–8\n",
"# extra code – this cell generates and saves Figure 6–8\n",
"\n",
"plt.figure(figsize=(8, 4))\n",
"\n",
@ -693,7 +693,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 6–9\n",
"# extra code – this cell generates and saves Figure 6–9\n",
"\n",
"plt.figure(figsize=(8, 4))\n",
"y_pred = tree_clf_tweaked.predict(X_iris_all).reshape(lengths.shape)\n",

24
07_ensemble_learning_and_random_forests.ipynb
View File

@ -133,7 +133,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 7–3\n",
"# extra code – this cell generates and saves Figure 7–3\n",
"\n",
"import matplotlib.pyplot as plt\n",
"import numpy as np\n",
@ -273,7 +273,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 7–5\n",
"# extra code – this cell generates and saves Figure 7–5\n",
"\n",
"def plot_decision_boundary(clf, X, y, alpha=1.0):\n",
" axes=[-1.5, 2.4, -1, 1.5]\n",
@ -363,7 +363,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code shows how to compute the 63% proba\n",
"# extra code – shows how to compute the 63% proba\n",
"print(1 - (1 - 1 / 1000) ** 1000)\n",
"print(1 - np.exp(-1))"
]
@ -413,7 +413,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code verifies that the predictions are identical\n",
"# extra code – verifies that the predictions are identical\n",
"bag_clf.fit(X_train, y_train)\n",
"y_pred_bag = bag_clf.predict(X_test)\n",
"np.all(y_pred_bag == y_pred_rf) # same predictions"
@ -447,7 +447,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 7–6\n",
"# extra code – this cell generates and saves Figure 7–6\n",
"\n",
"from sklearn.datasets import fetch_openml\n",
"\n",
@ -480,7 +480,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 7–8\n",
"# extra code – this cell generates and saves Figure 7–8\n",
"\n",
"m = len(X_train)\n",
"\n",
@ -534,8 +534,8 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – in case you're curious to see what the decision boundary\n",
"# looks like for the AdaBoost classifier\n",
"# extra code – in case you're curious to see what the decision boundary\n",
"# looks like for the AdaBoost classifier\n",
"plot_decision_boundary(ada_clf, X_train, y_train)"
]
},
@ -615,7 +615,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 7–9\n",
"# extra code – this cell generates and saves Figure 7–9\n",
"\n",
"def plot_predictions(regressors, X, y, axes, style,\n",
" label=None, data_style=\"b.\", data_label=None):\n",
@ -715,7 +715,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 7–10\n",
"# extra code – this cell generates and saves Figure 7–10\n",
"\n",
"fix, axes = plt.subplots(ncols=2, figsize=(10, 4), sharey=True)\n",
"\n",
@ -743,7 +743,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – at least not in this chapter, it's presented in chapter 2\n",
"# extra code – at least not in this chapter, it's presented in chapter 2\n",
"\n",
"import tarfile\n",
"import urllib.request\n",
@ -795,7 +795,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – evaluate the RMSE stats for the hgb_reg model\n",
"# extra code – evaluate the RMSE stats for the hgb_reg model\n",
"\n",
"from sklearn.model_selection import cross_val_score\n",
"\n",

40
08_dimensionality_reduction.ipynb
View File

@ -91,8 +91,8 @@
"plt.rc('font', size=14)\n",
"plt.rc('axes', labelsize=14, titlesize=14)\n",
"plt.rc('legend', fontsize=14)\n",
"plt.rc('xtick',labelsize=10)\n",
"plt.rc('ytick',labelsize=10)"
"plt.rc('xtick', labelsize=10)\n",
"plt.rc('ytick', labelsize=10)"
]
},
{
@ -147,7 +147,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book\n",
"# extra code\n",
"\n",
"import numpy as np\n",
"from scipy.spatial.transform import Rotation\n",
@ -177,7 +177,7 @@
},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 8–2\n",
"# extra code – this cell generates and saves Figure 8–2\n",
"\n",
"import matplotlib.pyplot as plt\n",
"from mpl_toolkits.mplot3d import Axes3D\n",
@ -245,7 +245,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 8–3\n",
"# extra code – this cell generates and saves Figure 8–3\n",
"\n",
"fig = plt.figure()\n",
"ax = fig.add_subplot(1, 1, 1, aspect='equal')\n",
@ -281,7 +281,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 8–4\n",
"# extra code – this cell generates and saves Figure 8–4\n",
"\n",
"from matplotlib.colors import ListedColormap\n",
"\n",
@ -305,7 +305,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves plots for Figure 8–5\n",
"# extra code – this cell generates and saves plots for Figure 8–5\n",
"\n",
"plt.figure(figsize=(10, 4))\n",
"\n",
@ -332,7 +332,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves plots for Figure 8–6\n",
"# extra code – this cell generates and saves plots for Figure 8–6\n",
" \n",
"axes = [-11.5, 14, -2, 23, -12, 15]\n",
"x2s = np.linspace(axes[2], axes[3], 10)\n",
@ -404,7 +404,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 8–7\n",
"# extra code – this cell generates and saves Figure 8–7\n",
"\n",
"angle = np.pi / 5\n",
"stretch = 5\n",
@ -507,7 +507,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code shows how to construct Σ from s\n",
"# extra code – shows how to construct Σ from s\n",
"m, n = X.shape\n",
"Σ = np.zeros_like(X_centered)\n",
"Σ[:n, :n] = np.diag(s)\n",
@ -616,7 +616,7 @@
"metadata": {},
"outputs": [],
"source": [
"1 - pca.explained_variance_ratio_.sum() # not in the book"
"1 - pca.explained_variance_ratio_.sum() # extra code"
]
},
{
@ -678,7 +678,7 @@
"metadata": {},
"outputs": [],
"source": [
"pca.explained_variance_ratio_.sum() # not in the book"
"pca.explained_variance_ratio_.sum() # extra code"
]
},
{
@ -694,7 +694,7 @@
"metadata": {},
"outputs": [],
"source": [
"plt.figure(figsize=(6,4))\n",
"plt.figure(figsize=(6, 4))\n",
"plt.plot(cumsum, linewidth=3)\n",
"plt.axis([0, 400, 0, 1])\n",
"plt.xlabel(\"Dimensions\")\n",
@ -795,7 +795,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 8–9\n",
"# extra code – this cell generates and saves Figure 8–9\n",
"\n",
"plt.figure(figsize=(7, 4))\n",
"for idx, X in enumerate((X_train[::2100], X_recovered[::2100])):\n",
@ -929,7 +929,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – show the equation computed by johnson_lindenstrauss_min_dim\n",
"# extra code – show the equation computed by johnson_lindenstrauss_min_dim\n",
"d = int(4 * np.log(m) / (ε ** 2 / 2 - ε ** 3 / 3))\n",
"d"
]
@ -983,7 +983,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – performance comparison between Gaussian and Sparse RP\n",
"# extra code – performance comparison between Gaussian and Sparse RP\n",
"\n",
"from sklearn.random_projection import SparseRandomProjection\n",
"\n",
@ -1027,7 +1027,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 8–10\n",
"# extra code – this cell generates and saves Figure 8–10\n",
"\n",
"plt.scatter(X_unrolled[:, 0], X_unrolled[:, 1],\n",
" c=t, cmap=darker_hot)\n",
@ -1047,7 +1047,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – shows how well correlated z1 is to t: LLE worked fine\n",
"# extra code – shows how well correlated z1 is to t: LLE worked fine\n",
"plt.title(\"$z_1$ vs $t$\")\n",
"plt.scatter(X_unrolled[:, 0], t, c=t, cmap=darker_hot)\n",
"plt.xlabel(\"$z_1$\")\n",
@ -1099,11 +1099,11 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 8–11\n",
"# extra code – this cell generates and saves Figure 8–11\n",
"\n",
"titles = [\"MDS\", \"Isomap\", \"t-SNE\"]\n",
"\n",
"plt.figure(figsize=(11,4))\n",
"plt.figure(figsize=(11, 4))\n",
"\n",
"for subplot, title, X_reduced in zip((131, 132, 133), titles,\n",
" (X_reduced_mds, X_reduced_isomap, X_reduced_tsne)):\n",

90
09_unsupervised_learning.ipynb
View File

@ -91,8 +91,8 @@
"plt.rc('font', size=14)\n",
"plt.rc('axes', labelsize=14, titlesize=14)\n",
"plt.rc('legend', fontsize=14)\n",
"plt.rc('xtick',labelsize=10)\n",
"plt.rc('ytick',labelsize=10)"
"plt.rc('xtick', labelsize=10)\n",
"plt.rc('ytick', labelsize=10)"
]
},
{
@ -147,7 +147,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 9–1\n",
"# extra code – this cell generates and saves Figure 9–1\n",
"\n",
"import matplotlib.pyplot as plt\n",
"from sklearn.datasets import load_iris\n",
@ -192,7 +192,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book\n",
"# extra code\n",
"\n",
"import numpy as np\n",
"from scipy import stats\n",
@ -263,7 +263,7 @@
"from sklearn.cluster import KMeans\n",
"from sklearn.datasets import make_blobs\n",
"\n",
"# not in the book – the exact arguments of make_blobs() are not important\n",
"# extra code – the exact arguments of make_blobs() are not important\n",
"blob_centers = np.array([[ 0.2, 2.3], [-1.5 , 2.3], [-2.8, 1.8],\n",
" [-2.8, 2.8], [-2.8, 1.3]])\n",
"blob_std = np.array([0.4, 0.3, 0.1, 0.1, 0.1])\n",
@ -288,7 +288,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 9–2\n",
"# extra code – this cell generates and saves Figure 9–2\n",
"\n",
"def plot_clusters(X, y=None):\n",
" plt.scatter(X[:, 0], X[:, 1], c=y, s=1)\n",
@ -399,7 +399,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 9–3\n",
"# extra code – this cell generates and saves Figure 9–3\n",
"\n",
"def plot_data(X):\n",
" plt.plot(X[:, 0], X[:, 1], 'k.', markersize=2)\n",
@ -489,7 +489,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book\n",
"# extra code\n",
"np.linalg.norm(np.tile(X_new, (1, k)).reshape(-1, k, 2)\n",
" - kmeans.cluster_centers_, axis=2).round(2)"
]
@ -532,7 +532,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 9–4\n",
"# extra code – this cell generates and saves Figure 9–4\n",
"\n",
"kmeans_iter1 = KMeans(n_clusters=5, init=\"random\", n_init=1, max_iter=1,\n",
" random_state=5)\n",
@ -600,7 +600,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 9–5\n",
"# extra code – this cell generates and saves Figure 9–5\n",
"\n",
"def plot_clusterer_comparison(clusterer1, clusterer2, X, title1=None,\n",
" title2=None):\n",
@ -647,7 +647,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book\n",
"# extra code\n",
"plt.figure(figsize=(8, 4))\n",
"plot_decision_boundaries(kmeans, X)"
]
@ -681,7 +681,7 @@
"metadata": {},
"outputs": [],
"source": [
"kmeans_rnd_init1.inertia_ # not in the book"
"kmeans_rnd_init1.inertia_ # extra code"
]
},
{
@ -690,7 +690,7 @@
"metadata": {},
"outputs": [],
"source": [
"kmeans_rnd_init2.inertia_ # not in the book"
"kmeans_rnd_init2.inertia_ # extra code"
]
},
{
@ -706,7 +706,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book\n",
"# extra code\n",
"X_dist = kmeans.transform(X)\n",
"(X_dist[np.arange(len(X_dist)), kmeans.labels_] ** 2).sum()"
]
@ -754,7 +754,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book\n",
"# extra code\n",
"kmeans_rnd_10_inits = KMeans(n_clusters=5, init=\"random\", n_init=10,\n",
" random_state=2)\n",
"kmeans_rnd_10_inits.fit(X)"
@ -773,7 +773,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book\n",
"# extra code\n",
"plt.figure(figsize=(8, 4))\n",
"plot_decision_boundaries(kmeans_rnd_10_inits, X)\n",
"plt.show()"
@ -964,7 +964,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 9–6\n",
"# extra code – this cell generates and saves Figure 9–6\n",
"\n",
"from timeit import timeit\n",
"\n",
@ -981,7 +981,7 @@
" inertias[k - 1, 0] = kmeans_.inertia_\n",
" inertias[k - 1, 1] = minibatch_kmeans.inertia_\n",
"\n",
"plt.figure(figsize=(10,4))\n",
"plt.figure(figsize=(10, 4))\n",
"\n",
"plt.subplot(121)\n",
"plt.plot(range(1, max_k + 1), inertias[:, 0], \"r--\", label=\"K-Means\")\n",
@ -1024,7 +1024,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 9–7\n",
"# extra code – this cell generates and saves Figure 9–7\n",
"\n",
"kmeans_k3 = KMeans(n_clusters=3, random_state=42)\n",
"kmeans_k8 = KMeans(n_clusters=8, random_state=42)\n",
@ -1072,7 +1072,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 9–8\n",
"# extra code – this cell generates and saves Figure 9–8\n",
"\n",
"kmeans_per_k = [KMeans(n_clusters=k, random_state=42).fit(X)\n",
" for k in range(1, 10)]\n",
@ -1104,7 +1104,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book\n",
"# extra code\n",
"plot_decision_boundaries(kmeans_per_k[4 - 1], X)\n",
"plt.show()"
]
@ -1147,7 +1147,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 9–9\n",
"# extra code – this cell generates and saves Figure 9–9\n",
"\n",
"silhouette_scores = [silhouette_score(X, model.labels_)\n",
" for model in kmeans_per_k[1:]]\n",
@ -1182,7 +1182,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 9–10\n",
"# extra code – this cell generates and saves Figure 9–10\n",
"\n",
"from sklearn.metrics import silhouette_samples\n",
"from matplotlib.ticker import FixedLocator, FixedFormatter\n",
@ -1253,7 +1253,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 9–11\n",
"# extra code – this cell generates and saves Figure 9–11\n",
"\n",
"X1, y1 = make_blobs(n_samples=1000, centers=((4, -4), (0, 0)), random_state=42)\n",
"X1 = X1.dot(np.array([[0.374, 0.95], [0.732, 0.598]]))\n",
@ -1303,7 +1303,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book\n",
"# extra code\n",
"\n",
"root = \"https://raw.githubusercontent.com/ageron/handson-ml3/main/\"\n",
"filename = \"ladybug.png\"\n",
@ -1344,7 +1344,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 9–12\n",
"# extra code – this cell generates and saves Figure 9–12\n",
"\n",
"segmented_imgs = []\n",
"n_colors = (10, 8, 6, 4, 2)\n",
@ -1353,7 +1353,7 @@
" segmented_img = kmeans.cluster_centers_[kmeans.labels_]\n",
" segmented_imgs.append(segmented_img.reshape(image.shape))\n",
"\n",
"plt.figure(figsize=(10,5))\n",
"plt.figure(figsize=(10, 5))\n",
"plt.subplots_adjust(wspace=0.05, hspace=0.1)\n",
"\n",
"plt.subplot(2, 3, 1)\n",
@ -1367,7 +1367,7 @@
" plt.title(f\"{n_clusters} colors\")\n",
" plt.axis('off')\n",
"\n",
"save_fig('image_segmentation_diagram', tight_layout=False)\n",
"save_fig('image_segmentation_plot', tight_layout=False)\n",
"plt.show()"
]
},
@ -1440,7 +1440,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – measure the accuracy when we use the whole training set\n",
"# extra code – measure the accuracy when we use the whole training set\n",
"log_reg_full = LogisticRegression(max_iter=10_000)\n",
"log_reg_full.fit(X_train, y_train)\n",
"log_reg_full.score(X_test, y_test)"
@ -1479,7 +1479,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 9–13\n",
"# extra code – this cell generates and saves Figure 9–13\n",
"\n",
"plt.figure(figsize=(8, 2))\n",
"for index, X_representative_digit in enumerate(X_representative_digits):\n",
@ -1488,7 +1488,7 @@
" interpolation=\"bilinear\")\n",
" plt.axis('off')\n",
"\n",
"save_fig(\"representative_images_diagram\", tight_layout=False)\n",
"save_fig(\"representative_images_plot\", tight_layout=False)\n",
"plt.show()"
]
},
@ -1694,7 +1694,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 9–14\n",
"# extra code – this cell generates and saves Figure 9–14\n",
"\n",
"def plot_dbscan(dbscan, X, size, show_xlabels=True, show_ylabels=True):\n",
" core_mask = np.zeros_like(dbscan.labels_, dtype=bool)\n",
@ -1747,7 +1747,7 @@
"metadata": {},
"outputs": [],
"source": [
"dbscan = dbscan2 # not in the book – the text says we now use eps=0.2"
"dbscan = dbscan2 # extra code – the text says we now use eps=0.2"
]
},
{
@ -1787,7 +1787,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 9–15\n",
"# extra code – this cell generates and saves Figure 9–15\n",
"\n",
"plt.figure(figsize=(6, 3))\n",
"plot_decision_boundaries(knn, X, show_centroids=False)\n",
@ -2172,7 +2172,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – bonus material\n",
"# extra code – bonus material\n",
"\n",
"resolution = 100\n",
"grid = np.arange(-10, 10, 1 / resolution)\n",
@ -2197,7 +2197,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cells generates and saves Figure 9–16\n",
"# extra code – this cells generates and saves Figure 9–16\n",
"\n",
"from matplotlib.colors import LogNorm\n",
"\n",
@ -2256,7 +2256,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 9–17\n",
"# extra code – this cell generates and saves Figure 9–17\n",
"\n",
"gm_full = GaussianMixture(n_components=3, n_init=10,\n",
" covariance_type=\"full\", random_state=42)\n",
@ -2294,7 +2294,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – comparing covariance_type=\"full\" and covariance_type=\"diag\"\n",
"# extra code – comparing covariance_type=\"full\" and covariance_type=\"diag\"\n",
"compare_gaussian_mixtures(gm_full, gm_diag, X)\n",
"plt.tight_layout()\n",
"plt.show()"
@ -2331,7 +2331,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 9–18\n",
"# extra code – this cell generates and saves Figure 9–18\n",
"\n",
"plt.figure(figsize=(8, 4))\n",
"\n",
@ -2373,7 +2373,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 9–19\n",
"# extra code – this cell generates and saves Figure 9–19\n",
"\n",
"from scipy.stats import norm\n",
"\n",
@ -2387,7 +2387,7 @@
"stds = np.linspace(stds_range[0], stds_range[1], 501)\n",
"Xs, Stds = np.meshgrid(xs, stds)\n",
"Z = 2 * norm.pdf(Xs - 1.0, 0, Stds) + norm.pdf(Xs + 4.0, 0, Stds)\n",
"Z = Z / Z.sum(axis=1)[:,np.newaxis] / (xs[1] - xs[0])\n",
"Z = Z / Z.sum(axis=1)[:, np.newaxis] / (xs[1] - xs[0])\n",
"\n",
"x_example_idx = (xs >= x_val).argmax() # index of the first value >= x_val\n",
"max_idx = Z[:, x_example_idx].argmax()\n",
@ -2479,7 +2479,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – bonus material\n",
"# extra code – bonus material\n",
"n_clusters = 3\n",
"n_dims = 2\n",
"n_params_for_weights = n_clusters - 1\n",
@ -2514,7 +2514,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 9–20\n",
"# extra code – this cell generates and saves Figure 9–20\n",
"\n",
"gms_per_k = [GaussianMixture(n_components=k, n_init=10, random_state=42).fit(X)\n",
" for k in range(1, 11)]\n",
@ -2576,7 +2576,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this figure is almost identical to Figure 9–16\n",
"# extra code – this figure is almost identical to Figure 9–16\n",
"plt.figure(figsize=(8, 5))\n",
"plot_gaussian_mixture(bgm, X)\n",
"plt.show()"
@ -2588,7 +2588,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 9–21\n",
"# extra code – this cell generates and saves Figure 9–21\n",
"\n",
"X_moons, y_moons = make_moons(n_samples=1000, noise=0.05, random_state=42)\n",
"\n",

60
10_neural_nets_with_keras.ipynb
View File

@ -77,7 +77,7 @@
"cell_type": "markdown",
"metadata": {},
"source": [
"And TensorFlow ≥ 2.6:"
"And TensorFlow ≥ 2.7:"
]
},
{
@ -88,7 +88,7 @@
"source": [
"import tensorflow as tf\n",
"\n",
"assert tf.__version__ >= \"2.6.0\""
"assert tf.__version__ >= \"2.7.0\""
]
},
{
@ -189,7 +189,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – extra material\n",
"# extra code – shows how to build and train a Perceptron\n",
"\n",
"from sklearn.linear_model import SGDClassifier\n",
"\n",
@ -213,7 +213,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – extra material\n",
"# extra code – plots the decision boundary of a Perceptron on the iris dataset\n",
"\n",
"import matplotlib.pyplot as plt\n",
"from matplotlib.colors import ListedColormap\n",
@ -256,7 +256,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code generates and saves Figure 10–8\n",
"# extra code – this cell generates and saves Figure 10–8\n",
"\n",
"from scipy.special import expit as sigmoid\n",
"\n",
@ -358,7 +358,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this was left as an exercise for the reader\n",
"# extra code – this was left as an exercise for the reader\n",
"\n",
"from sklearn.datasets import load_iris\n",
"from sklearn.model_selection import train_test_split\n",
@ -470,7 +470,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book\n",
"# extra code\n",
"\n",
"plt.imshow(X_train[0], cmap=\"binary\")\n",
"plt.axis('off')\n",
@ -539,7 +539,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 10–10\n",
"# extra code – this cell generates and saves Figure 10–10\n",
"\n",
"n_rows = 4\n",
"n_cols = 10\n",
@ -585,7 +585,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – clear the session to reset the name counters\n",
"# extra code – clear the session to reset the name counters\n",
"tf.keras.backend.clear_session()\n",
"tf.random.set_seed(42)\n",
"\n",
@ -612,7 +612,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – another way to display the model's architecture\n",
"# extra code – another way to display the model's architecture\n",
"tf.keras.utils.plot_model(model, \"my_fashion_mnist_model.png\", show_shapes=True)"
]
},
@ -712,7 +712,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell is equivalent to the previous cell\n",
"# extra code – this cell is equivalent to the previous cell\n",
"model.compile(loss=tf.keras.losses.sparse_categorical_crossentropy,\n",
" optimizer=tf.keras.optimizers.SGD(),\n",
" metrics=[tf.keras.metrics.sparse_categorical_accuracy])"
@ -724,7 +724,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code shows how to convert class ids to one-hot vectors\n",
"# extra code – shows how to convert class ids to one-hot vectors\n",
"tf.keras.utils.to_categorical([0, 5, 1, 0], num_classes=10)"
]
},
@ -741,7 +741,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this code shows how to convert one-hot vectors to class ids\n",
"# extra code – shows how to convert one-hot vectors to class ids\n",
"np.argmax(\n",
" [[1., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
" [0., 0., 0., 0., 0., 1., 0., 0., 0., 0.],\n",
@ -798,8 +798,8 @@
"pd.DataFrame(history.history).plot(\n",
" figsize=(8, 5), xlim=[0, 29], ylim=[0, 1], grid=True, xlabel=\"Epoch\",\n",
" style=[\"r--\", \"r--.\", \"b-\", \"b-*\"])\n",
"plt.legend(loc=\"lower left\") # not in the book\n",
"save_fig(\"keras_learning_curves_plot\") # not in the book\n",
"plt.legend(loc=\"lower left\") # extra code\n",
"save_fig(\"keras_learning_curves_plot\") # extra code\n",
"plt.show()"
]
},
@ -809,7 +809,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – shows how to shift the training curve by -1/2 epoch\n",
"# extra code – shows how to shift the training curve by -1/2 epoch\n",
"plt.figure(figsize=(8, 5))\n",
"for key, style in zip(history.history, [\"r--\", \"r--.\", \"b-\", \"b-*\"]):\n",
" epochs = np.array(history.epoch) + (0 if key.startswith(\"val_\") else -0.5)\n",
@ -883,7 +883,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – this cell generates and saves Figure 10–12\n",
"# extra code – this cell generates and saves Figure 10–12\n",
"plt.figure(figsize=(7.2, 2.4))\n",
"for index, image in enumerate(X_new):\n",
" plt.subplot(1, 3, index + 1)\n",
@ -915,7 +915,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – load and split the California housing dataset, like earlier\n",
"# extra code – load and split the California housing dataset, like earlier\n",
"housing = fetch_california_housing()\n",
"X_train_full, X_test, y_train_full, y_test = train_test_split(\n",
" housing.data, housing.target, random_state=42)\n",
@ -986,7 +986,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – reset the name counters and make the code reproducible\n",
"# extra code – reset the name counters and make the code reproducible\n",
"tf.keras.backend.clear_session()\n",
"tf.random.set_seed(42)"
]
@ -1050,7 +1050,7 @@
"metadata": {},
"outputs": [],
"source": [
"tf.random.set_seed(42) # not in the book"
"tf.random.set_seed(42) # extra code"
]
},
{
@ -1220,7 +1220,7 @@
" aux_output = self.aux_output(hidden2)\n",
" return output, aux_output\n",
"\n",
"tf.random.set_seed(42) # not in the book – just for reproducibility\n",
"tf.random.set_seed(42) # extra code – just for reproducibility\n",
"model = WideAndDeepModel(30, activation=\"relu\", name=\"my_cool_model\")"
]
},
@ -1256,7 +1256,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – delete the directory, in case it already exists\n",
"# extra code – delete the directory, in case it already exists\n",
"\n",
"import shutil\n",
"\n",
@ -1278,7 +1278,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – show the contents of the my_keras_model/ directory\n",
"# extra code – show the contents of the my_keras_model/ directory\n",
"for path in sorted(Path(\"my_keras_model\").glob(\"**/*\")):\n",
" print(path)"
]
@ -1317,7 +1317,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – show the list of my_weights.* files\n",
"# extra code – show the list of my_weights.* files\n",
"for path in sorted(Path().glob(\"my_weights.*\")):\n",
" print(path)"
]
@ -1335,7 +1335,7 @@
"metadata": {},
"outputs": [],
"source": [
"shutil.rmtree(\"my_checkpoints\", ignore_errors=True) # not in the book"
"shutil.rmtree(\"my_checkpoints\", ignore_errors=True) # extra code"
]
},
{
@ -1411,7 +1411,7 @@
"metadata": {},
"outputs": [],
"source": [
"if \"google.colab\" in sys.modules: # not in the book\n",
"if \"google.colab\" in sys.modules: # extra code\n",
" %pip install -q -U tensorboard-plugin-profile"
]
},
@ -1447,7 +1447,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – builds the first regression model we used earlier\n",
"# extra code – builds the first regression model we used earlier\n",
"tf.keras.backend.clear_session()\n",
"tf.random.set_seed(42)\n",
"norm_layer = tf.keras.layers.Normalization(input_shape=X_train.shape[1:])\n",
@ -1516,7 +1516,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book\n",
"# extra code\n",
"\n",
"if \"google.colab\" in sys.modules:\n",
" from google.colab import output\n",
@ -1574,7 +1574,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book\n",
"# extra code\n",
"\n",
"if \"google.colab\" in sys.modules:\n",
" !tensorboard dev upload --logdir ./my_logs --one_shot \\\n",
@ -1632,7 +1632,7 @@
"metadata": {},
"outputs": [],
"source": [
"# not in the book – extra material\n",
"# extra code – lists all running TensorBoard server instances\n",
"\n",
"from tensorboard import notebook\n",
"\n",